Pisias, N.G., Mayer, L.A., Janecek, T.R., Palmer-Julson, A., and van Andel, T.H. (Eds.), 1995Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 138

27. ALKENONE SEA-SURFACE TEMPERATURES AND CARBON BURIAL AT SITE 846(EASTERN EQUATORIAL PACIFIC OCEAN): THE LAST 1.3 M.Y.1

K.-C. Emeis,2 H. Doose,2 A. Mix,3 and D. Schulz-Bull4

ABSTRACT

We analyzed the unsaturation ratio (U k ) of long-chain ketones—a molecular sea-surface temperature (SST) indicator—con-centrations of carbonate and organic carbon in sediments from Site 846 in the eastern equatorial Pacific Ocean. Based on anisotopic age model for the composite depth section of 0-46 m below seafloor and on estimates of sediment density, accumulationrates of these biogenic compounds were calculated. Our combined temperature and biogenic flux record traces conditions at theorigin of the South Equatorial Current over the last 1.3 m.y.

S STs have fluctuated considerably over the interval studied. A long trend of gradual decrease from 24°C at 1.3 Ma ends between500 and 400 Ka, when lowest values of 19°C were reached. Since this time, the temperature data indicate a warming trend to theHolocene modulated by high-amplitude variation (19° to 27°C). The inversion of the trend between 400 and 500 Ka coincideswith maximal accumulation rates of carbonate, which since then have decreased. In contrast, organic carbon accumulation sincethen has increased in variability and in absolute values.

On shorter time scales, the records show a strong link to the global climatic background. Since 1.3 Ma, carbonate (0.2-3 g/cm2

k.y.~1) and organic carbon accumulation rates (2-30 mg/cm2 k.y.~1) were consistently high (more than twice their modern valuesand those of interglacials) during glacial maxima in the benthic isotope record, when temperatures were low.

However, cross-spectral analyses with the δ 1 8 θ record suggest that variation in organic carbon flux is not linked directly tovariations in SST. Temperature maxima in our record led interglacial events by 7 k.y. in the 100-k.y. eccentricity cycle and by 5k.y. in the 41-k.y. obliquity cycle. In contrast, maxima in organic carbon accumulation lag behind glacial maxima and lowtemperatures by 14 k.y. in the eccentricity cycle. On glacial/interglacial time scales, a prominent influence on SST—but not onorganic carbon burial—at Site 846 appears to be the advection of cold water into the South Equatorial Current.

INTRODUCTION

The primary objective of Ocean Drilling Program (ODP) Leg 138was to reconstruct the paleoceanographic history of the eastern equa-torial Pacific Ocean, an area that is characterized today by intensegradients in physical, chemical, and biological properties of surfacewaters. Among the diagnostic characteristics of each water massinteracting here are temperatures at the sea surface and biologicalproductivity in the euphotic zone (e.g., Wooster, 1969; Bryden andBrady, 1985; Chavez et al., 1990; Chavez and Barber, 1987).

Site 846 of Leg 138 (3°05.696'S and 90°49.078'W in a waterdepth of 3307 m) is situated 300 km south of Galapagos (Fig. 1). It islocated in the southeastern origin of the cold tongue of the EquatorialPacific Ocean (Wyrtki, 1981), where the Peru Current merges withthe South Equatorial Current (SEC). Here, combined eastward shoal-ing of the Equatorial Undercurrent (EUC) and advection of waterfrom the eastern boundary current result in high nutrient levels in theeuphotic zone, high biological production, and cold SSTs (Chavezand Barber, 1987; Halpern et al., 1990).

The extent of the cold tongue, which is defined by meridionaltemperature gradients of 0.5°C per 100 km north and south of theequator, is traced by the unsaturation ratio of long-chain ketones insurface sediments (Fluegge, unpubl. data), which are a robust andaccurate molecular thermometer (Prahl et al., 1988; Eglinton et al.,1992; Lyle et al., 1992). Today, temperatures at the site display apronounced annual cycle that ranges from 20° to 27°C (Halpern et al.,1990). SSTs of the climatological mean in the period from July toSeptember range from 21° to 22°C; modern observations indicate

Pisias, N.G., Mayer, L.A., Janecek, T.R., Palmer-Julson, A., and van Andel, T.H.(Eds.), 1995. Proc. ODP, Sci. Results, 138: College Station, TX (Ocean Drilling Program).

2 Geologisch-Palaontologisches Institut, Universitat Kiel, Federal Republic ofGermany.

3 School of Oceanography, Oregon State University, Corvallis, OR, U.S.A.4 Institut für Meereskunde, Universitat Kiel, Federal Republic of Germany.

significant interannual variability (2°C). During the main period ofannual plankton productivity in early austral winter (Dandonneau andEldin, 1987), temperatures at the site average 21° to 23°C and coin-cide with intensified southeast trade winds and maximal intensity ofthe Peru Current.

According to surveys by Chavez et al. (1990), dominant primaryproducing organisms in the eastern equatorial Pacific are blue greenalgae, picoplankton, dinoflagellates, and diatoms. Coccolithophores,which produce the alkenones used for temperature reconstructions, areamong the dominant phytoplankton taxa, but do not contribute signifi-cantly to primary productivity in the eastern equatorial Pacific Ocean.However, their abundance correlates well with measured chlorophyllconcentrations in surface waters during peak primary production(Chavez et al., 1990), and we expect that the alkenone unsaturationindex records temperatures at the time of highest annual productivity.This assumption is corroborated by a survey of the U3

k7 indexes in

core-top samples: Fluegge (unpubl. data) found an alkenone tempera-ture of 22.5°C close to the location of Site 846. His contour maps ofmodern alkenone SSTs in the area depict a strong temperature gradientover a range of 22° to 24°C, and the zone of low temperatures corre-sponds well with the area of highest TOC accumulation rates in core-top samples (Lyle, 1992).

Site 846 was located to explore the history of interaction betweenthe eastern boundary current and the SEC, both of which are stronglyinfluenced by changes in the atmospheric circulation of the SouthernHemisphere on seasonal, interannual, and geological time scales. Asa first contribution to this end, we analyzed concentrations and accu-mulation rates of organic carbon and carbonate, and the unsaturationratio of long-chain alkenones in lipid extracts of samples from the top46 m of the sedimentary record recovered at Site 846.

METHODS

Drilling four offset holes recovered a continuous and highly re-solved sediment record at Site 846 (Mayer, Pisias, Janecek, et al.,

605

K.-C. EMEIS, H. DOOSE, A. MIX, D. SCHULZ-BULL

10°

5°

o°

10°

15°

EquatorialUndercurrent

South- Equatorial

Current

100°

Site846

90°

Figure 1. Location of Site 846 in the eastern Pacific Ocean. Major currentsaffecting the surface hydrography at Site 846 are indicated. The cold tongueextends off South America and is marked by low SST and high nutrientconcentrations (after Wooster, 1969).

1992). Age assignments for our samples are based on their position inthe revised composite depth section (revised meters composite depth[rmcd]; Hagelberg et al., this volume) and the corresponding depth/agemodel of Mix et al. (this volume). Average sample interval of ouraccumulation rate time series is 5.5 k.y. (maximum of 24 k.y.), that ofSST is 6.5 k.y. (maximum of 24 k.y.). The mean sampling interval ofthe isotope record is 2.6 k.y. (maximum of 21 k.y.). To computeaccumulation rates of individual components from concentration data,we adopted linear sedimentation rates from Mix et al. (this volume) andinterpolated for each sample interval of our sample set.

Comparing the smoothed GRAPE wet-bulk density (WBD) rec-ord of the composite section with dry-bulk density (DBD) measure-ments on discrete samples (Mayer, Pisias, Janecek, et al., 1992), weestimated DBD of each sample in our set based on the correspondingGRAPE density value in the composite depth scale (Fig. 2) as

DBD (g/cm3) = -1.48 + 1.4 × GRAPE density. (1)

Mass accumulation rates (MAR) and accumulation rates of or-ganic carbon (TOC MAR) and carbonate (CaCO3 MAR) were calcu-lated as the product of concentration, linear sedimentation rate, andDBD estimates as follows:

MAR (g/cm2 k.y."1) = LSR (cm/k.y.) × DBD(g/cm3),

TOC MAR (mg/cm2 ley."1) = (MAR/100 × %TOC) × 1000,

CaCO3 MAR (g/cm2 k.y.-1) = MAR/100 × %CaCO3.

Total carbon (TC) concentrations were measured as CO2 in a CarloErba Model 1108 elemental analyzer after flash combustion of samplein a tin foil container at approximately 1600°C (oxidation column tem-perature 1020°C) in helium carrier gas, catalytic oxidation, chromato-graphic separation, and detection in a thermal conductivity detector.Calculation of response factors of standards (acetanilide; one standardand one blank run for each 10 sample runs) and corrections for blankspermitted us to calculate weight percentages of carbon and nitrogen inthe sample. Duplicate analyses of samples indicate that the standarddeviation of total carbon analyses is less than 0.1% (n = 62). Tomeasure organic carbon, we acidified a weighed split of sample (ap-proximately 10 mg) in a tin foil container with successively strongerphosphoric acid (maximum 17%) until no further CO2 evolved. Theacidified sample containers were then dried at 40°C in a desiccator andanalyzed before the tin foil containers disintegrated. All samples were

o.θ

0.6

0.4

0.2

DBD = -1.48 + 1.4GRAPE; r 0.807

1.3 1.4 1.5

GRAPE density (g/cm3)

1.6

Figure 2. Correlation between measured DBD of discrete samples and stackedGRAPE density of corresponding intervals (Mayer, Pisias, Janecek, et al., 1992).

run in duplicate. Standard deviations average 0.05% TOC. Concen-tration of CaCO3 content was calculated by assuming that the differ-ence between total and organic carbon is carbonate carbon:

% CaCO3 = (TC - TOC) × 8.3334. (2)

To reconstruct SSTs, we measured the unsaturation ratio of long-chain ketones produced by coccolithophores of the family Gephyro-capsaceae in lipid extracts of sediments (Brassell et al., 1986). Weighedsplits of dried and homogenized sediments (0.3-3 g) were extractedtwice in an ultrasonic bath for 40 min each. The solvent was 2 × 20 mLtwice-distilled CH2C12, spiked with a known amount of 5α-cholestaneas an internal recovery standard. After each extraction step, sampleswere centrifuged and the solvent was collected by pipette. The twolipid extracts were combined and lipid classes were separated chroma-tographically on commercial silica columns (conditioned with 5 mLCH2C12) by elution with 10 mL CH2C12. The eluant was dried in arotary evaporator and redissolved in 10,20, or 50 µL hexane, depend-ing on original sediment weight and TOC concentration. The samplein hexane (1 µL) was injected in split mode onto a Fisons Instrumentsgas chromatograph fitted with a 50-m glass capillary column (Ultra-1;i.d., 0.35 mm, film thickness, 0.52 mm) and a flame ionization detector.Hydrogen was used as carrier gas (column head pressure 140 kPa). Thetemperature program was 150° to 300°C at 107min, 300° to 330°C at2°C/mm, and isothermal at 33O°C for 18 min.

Alkenones were identified by comparing retention times withthose of synthetic standards (provided by G. Eglinton, Bristol). Peakareas were converted to the U^7 index, and SSTs calculated accordingto a calibration based on culture experiments (Prahl et al., 1988). Forthe sake of clarity and compatibility with other temperature proxyindicators, we report our results based on the U3

k

7 index as tempera-tures (Prahl et al., 1988):

SST (°C) = (C37:2/(C37:2 + C37:3) -0.039)/0.034. (3)

Results of duplicate analyses in = 16) indicate that the standarddeviation of temperatures calculated from U ^ is 0.15°C (range0-0.35°C).s

RESULTS

Table 1 is a list of samples and analytical results from Site 846.Samples are identified based on standard ODP sample identification(hole, core, section, interval), shipboard depth in meters below sea-floor, and a composite depth assigned to each sample based on thespliced records for all four holes drilled at Site 846. Figure 3 shows

606

Table 1. Analytical results of chemical analyses for samples of Site 846.

Core, section, Depth Rev Mean Meaninterval (cm) (mbsf) mcd Age LSR WBD DBD BAR TC CaCO3 CaCO3-AR TOC ±lσ TOC-AR U3

K

7 SST ±1

138-846-Al-1,5-7 0.06 0.07 2.2 5.9 1.36 0.41 2.4 8.9 66.0 1.6 1.03 0.02 25.3 0.894 25.1 0.1:Al-1,20-22 0.21 0.25 6.1 4.3 1.41 0.48 2.0 9.7 75.3 1.5 0.62 0.02 12.4 0.910 25.6Al-1,40-42 0.41 0.56 13.7 4.1 1.40 0.46 1.9 9.6 71.5 1.4 1.01 0.00 19.5Al-1,60-62 0.61 0.73 17.7 4.3 1.38 0.43 1.8 8.9 64.3 1.2 1.15 0.07 21.1 0.863 24.2Al-1,80-82 0.81 0.90 21.7 4.4 1.37 0.42 1.9 9.3 66.0 1.2 1.42 0.04 26.3 0.864 24.3Al-1,100-102 1.01 1.05 25.0 4.6 1.37 0.43 1.9 9.4 68.2 1.3 1.22 0.02 23.7 0.860 24.1Al-1, 120-122 1.21 1.19 28.0 4.8 1.35 0.40 1.9 8.9 62.5 1.2 1.37 0.01 26.1Al-1, 140-142 1.41 1.34 31.1 5.2 1.38 0.44 2.2 8.5 60.7 1.3 1.20 0.03 26.3 0.939 26.5Al-2,20-22 1.71 1.69 37.7 5.5 1.37 0.42 2.4 8.0 59.7 1.4 0.79 0.08 18.7 0.904 25.4A1-2,40-42 1.91 1.88 40.9 6.0 1.36 0.41 2.4 7.1 52.5 1.3 0.78 0.00 19.1Al-2,61-63 2.12 2.07 44.0 6.3 1.35 0.39 2.5 6.3 42.0 1.0 1.21 0.02 30.1 0.830 23.3Al-2,80-82 2.31 2.23 46.5 6.5 1.33 0.37 2.4 6.0 41.9 1.0 0.99 0.01 23.6 0.811 22.7Al-2, 100-102 2.51 2.42 49.4 6.5 1.32 0.35 2.3 6.3 46.3 1.1 0.78 0.06 18.0 0.820 23.0Al-2, 120-122 2.71 2.67 53.3 6.4 1.32 0.35 2.3 6.3 43.9 1.0 0.99 0.03 22.4 0.928 26.1Al-2, 140-142 2.91 2.89 56.8 6.0 1.33 0.37 2.2 6.2 43.9 1.0 0.91 0.01 20.4 0.833 23.4Al-3, 12-14 3.13 3.15 61.3 5.5 1.32 0.35 1.9 6.0 40.6 0.8 1.09 0.00 20.8 0.858 24.1Dl-1,20-22 4.21 3.32 64.5 5.2 1.33 0.36 1.8 5.5 39.8 0.7 0.75 0.03 13.5 0.816 22.9Al-3,35-37 3.36 3.35 65.2 4.7 1.31 0.33 1.6 5.7 36.1 0.6 1.38 0.00 22.2 0.852 23.9Al-3,50-52 3.51 3.45 67.3 4.4 1.35 0.39 1.8 5.6 35.1 0.6 1.40 0.03 24.8Al-3,70-72 3.71 3.59 70.6 4.0 1.31 0.34 1.4 4.4 31.8 0.4 0.59 0.06 8.1 0.834 23.4Dl-1,60-62 4.61 3.76 75.1 3.7 1.26 0.27 1.0 4.5 27.5 0.3 1.14 10.9 0.862 24.2Al-3,90-92 3.91 3.79 76.0 3.2 1.31 0.34 1.1 5.6 40.3 0.5 0.73 0.01 8.4 0.849 23.8Al-3,110-112 4.11 3.96 81.3 3.1 1.32 0.36 1.1 4.7 34.5 0.4 0.60 0.10 6.5 0.899 25.3Al-3, 130-132 4.31 4.10 86.1 2.8 1.30 0.32 0.9 4.7 33.0 0.3 0.79 0.04 7.2 0.795 22.2Dl-1,120-122 5.21 4.35 95.3 2.8 1.30 0.32 0.9 4.9 31.6 0.3 0.69 6.1Dl-1, 140-142 5.41 4.58 103.4 3.0 1.31 0.34 1.0 4.7 31.2 0.3 0.93 0.04 9.2 0.834 23.4Dl-2,20-22 5.71 4.93 114.3 3.4 1.31 0.33 1.2 5.3 31.9 0.4 1.46 17.1 0.830 23.3Dl-2,40-42 5.91 5.15 120.2 3.9 1.34 0.39 1.5 7.6 46.0 0.7 2.04 30.5Dl-2,60-62 6.11 5.35 125.2 4.1 1.36 0.41 1.7 6.6 48.5 0.8 0.74 0.02 12.6 0.877 24.6Dl-2,80-82 6.31 5.55 129.9 4.3 1.31 0.34 1.5 7.5 48.5 0.7 0.72 10.7 0.903 25.4Dl-2, 100-102 6.51 5.75 134.5 4.4 1.39 0.44 2.0 0.50 9.9 0.834 23.4Dl-2, 120-122 6.71 5.95 138.9 4.6 1.42 0.49 2.2 7.8 57.2 1.3 0.98 21.7 0.983 27.8Dl-2, 140-142 6.91 6.16 143.5 4.6 1.41 0.48 2.2 0.914 25.7Dl-3,20-22 7.21 6.46 149.9 4.7 1.41 0.48 2.3 6.6 49.1 1.1 0.69 0.04 15.8 0.782 21.9Dl-3,40-42 7.41 6.66 154.1 4.8 1.42 0.49 2.4 7.2 53.0 1.3 0.87 0.03 20.6Dl-3,60-62 7.61 6.84 157.8 4.9 1.39 0.45 2.2 7.6 56.7 1.2 0.84 0.06 18.4 0.783 21.9Dl-3,80-82 7.81 7.03 161.7 4.8 1.37 0.42 2.0 7.0 50.4 1.0 0.98 0.02 19.9 0.794 22.2Dl-3, 100-102 8.01 7.21 165.5 4.7 1.40 0.46 2.2 1.00 21.9Dl-3, 120-122 8.21 7.39 169.4 4.4 1.38 0.44 2.0 7.7 57.5 1.1 0.80 0.01 15.6 0.774 21.6Dl-3, 140-142 8.43 7.58 173.9 3.8 1.37 0.42 1.6 6.5 45.7 0.8 1.05 0.01 17.4 0.989 27.9Dl-4,20-22 8.71 7.81 180.4 3.2 1.37 0.42 1.3 1.14 0.14 15.3 0.840 23.6Dl-4,40-42 8.91 7.98 186.3 2.6 1.35 0.39 1.0 7.3 53.6 0.5 0.84 0.02 8.5Dl-4,60-62 9.11 8.13 192.6 2.2 1.37 0.43 0.9 7.5 55.6 0.5 0.83 0.01 7.8 0.813 22.8Dl-4,80-82 9.31 8.29 200.7 1.9 1.41 0.47 0.9 7.3 56.0 0.5 0.62 0.03 5.5 0.875 24.6Dl-4, 100-102 9.51 8.44 209.1 1.8 1.40 0.47 0.8 0.42 0.01 3.5 0.857 24.1Dl-4, 120-122 9.71 8.61 218.1 2.2 1.46 0.55 1.2 7.9 60.9 0.7 0.55 0.00 6.4 0.849 23.8Dl-4, 144-146 9.95 8.84 227.4 3.0 1.44 0.51 1.6 8.3 62.6 1.0 0.82 0.06 12.7 0.847 23.8Dl-5,20-22 10.21 9.11 234.8 4.1 1.45 0.53 2.3 0.73 0.03 17.1 0.786 22.0Dl-5,40-42 10.41 9.32 239.0 5.4 1.41 0.48 2.7 7.4 56.7 1.5 0.61 0.02 16.2Dl-5,60-62 10.61 9.52 242.4 6.0 1.40 0.46 2.9 7.6 56.8 1.6 0.75 0.03 21.4 0.890 25.0Dl-5,80-82 10.81 9.71 245.5 5.9 1.43 0.51 3.0 8.1 61.1 1.9 0.76 0.05 23.0 0.909 25.6Dl-5, 100-102 11.01 9.90 248.8 5.8 1.47 0.56 3.2 0.77 0.04 25.0 0.880 24.7Dl-5, 120-122 11.21 10.09 252.1 5.6 1.46 0.55 3.1 9.2 71.0 2.2 0.68 0.08 21.4 0.768 21.4Dl-5, 140-142 11.41 10.30 256.0 4.5 1.43 0.50 2.5 7.6 57.8 1.4 0.68 0.01 16.8 0.710 19.7Dl-6,20-22 11.71 10.64 264.3 3.8 1.42 0.49 1.8 9.2 67.7 1.2 1.04 0.13 18.5 0.739 20.6Dl-6,40-42 11.91 10.87 271.1 3.3 1.39 0.46 1.5 0.88 0.01 12.8Dl-6,60-62 12.11 11.07 277.5 3.1 1.38 0.44 1.4 7.8 58.6 0.8 0.79 0.01 10.7 0.952 26.9Dl-6,80-82 12.31 11.26 283.7 3.1 1.44 0.52 1.6 8.8 68.9 1.1 0.53 0.05 8.5 0.817 22.9Dl-6, 100-102 12.51 11.44 289.4 3.2 1.40 0.47 1.5 0.78 0.09 11.6 0.820 23.0Dl-6, 120-122 12.71 11.62 295.0 3.2 1.44 0.52 1.7 8.3 63.5 1.1 0.65 0.00 11.0 0.771 21.5Dl-6, 140-142 12.91 11.82 301.1 3.3 1.47 0.57 1.9 8.2 58.8 1.1 1.14 0.07 21.2 0.806 22.6B2-3, 142-144 11.43 11.99 306.2 3.3 1.47 0.57 1.9 9.3 72.5 1.4 0.55 0.30 10.3 0.805 22.5B2-4,8-10 11.59 12.15 311.0 3.3 1.47 0.57 1.9 8.8 67.1 1.3 0.74 0.22 13.9 0.811 22.7C2-1,93-95 12.94 12.41 318.9 3.3 1.41 0.47 1.6 8.6 66.2 1.0 0.63 0.01 10.0 0.778 21.7C2-1, 110-112 13.11 12.54 322.7 3.5 1.48 0.58 2.0 9.2 72.0 1.4 0.55 0.04 10.8 0.806 22.6C2-1, 122-124 13.23 12.69 327.0 3.9 1.48 0.57 2.1 9.6 75.9 1.6 0.49 0.03 10.2 0.862 24.2C2-1, 140-142 13.41 13.10 337.1 4.2 1.45 0.53 2.4 9.6 73.9 1.8 0.68 0.03 16.5 0.879 24.7C2-2, 10-12 13.61 13.29 341.2 4.6 1.46 0.55 2.5 10.0 76.2 1.9 0.85 0.09 21.4 0.773 21.6C2-2,30-32 13.81 13.41 343.9 4.6 1.48 0.58 2.7 10.0 75.8 2.0 0.40 0.04 10.6 0.752 21.0C2-2,50-52 14.01 13.60 348.0 4.5 1.43 0.50 2.3 9.8 76.8 1.7 0.56 0.03 12.8 0.783 21.9C2-2,70-72 14.21 13.80 352.5 4.4 1.48 0.58 2.5 9.9 77.3 1.9 0.58 0.00 14.6 0.789 22.1C2-2,90-92 14.41 13.92 355.3 4.2 1.49 0.59 2.5 9.6 76.7 1.9 0.41 0.12 10.1 0.791 22.1C2-2, 110-112 14.61 14.03 358.0 4.0 1.47 0.56 2.3 9.9 76.1 1.8 0.80 0.02 18.5 0.755 21.1C2-2, 130-132 14.81 14.21 362.5 3.8 1.50 0.60 2.3 9.4 71.6 1.7 0.81 0.07 18.7C2-3, 1-3 15.02 14.52 371.0 3.5 1.48 0.58 2.0 9.4 74.8 1.5 0.46 0.01 9.3 0.747 20.8C2-3,23-25 15.24 14.74 377.6 3.3 1.52 0.63 2.0 9.8 75.3 1.5 0.78 0.05 15.9 0.765 21.4C2-3,42-44 15.43 14.91 382.9 3.2 1.49 0.59 1.9 10.1 77.7 1.4 0.50 0.11 9.3 0.930 26.2C2-3,62-64 15.63 15.12 389.6 3.2 1.48 0.57 1.8 9.5 0.31 0.05 5.6 0.837 23.5C2-3,83-85 15.84 15.41 398.4 3.3 1.50 0.61 2.1 10.2 0.47 0.29 9.7 0.813 22.8D2-1, 10-12 13.61 15.45 399.6 3.5 1.51 0.62 2.1 9.3 73.5 1.6 0.48 0.01 10.2 0.796 22.3D2-1,30-32 13.81 15.66 405.6 3.6 1.53 0.65 2.3 9.7 76.6 1.8 0.55 0.10 12.7D2-1,50-52 14.01 15.88 411.7 3.6 1.55 0.68 2.4 9.6 75.4 1.8 0.53 0.03 13.1D2-1,70-72 14.21 16.09 417.5 3.6 1.54 0.66 2.4 9.7 77.0 1.8 0.44 0.06 10.4 0.851 23.9D2-2, 10-12 14.54 16.44 427.3 3.6 1.59 0.73 2.6 9.4 72.9 1.9 0.63 0.01 16.2 0.783 21.9D2-2,30-32 14.74 16.64 432.9 3.6 1.54 0.65 2.4 10.0 67.0 1.6 1.79 0.29 42.4

K.-C. EMEIS, H. DOOSE, A. MIX, D. SCHULZ-BULL

Table 1 (continued).

Core, section,interval (cm)

D2-2, 50-52D2-2, 70-72D2-2, 90-92D2-2, 110-112D2-2, 130-132D2-3, 10-12D2-3, 30-32D2-3, 50-52D2-3, 70-72D2-3, 90-92D2-3, 110-112D2-4, 10-12D2-4, 30-32D2-4, 50-52D2-4, 70-72D2-4, 90-92D2-4, 110-112D2-4, 130-132D2-5, 10-12D2-5, 30-32D2-5, 50-52D2-5,70-72D2-5, 90-92D2-5, 110-112D2-5, 130-132D2-6, 10-12D2-6, 30-32D2-6, 50-52D2-6, 70-72D2-6, 90-92D2-6, 110-112D2-6, 130-132B3-4, 8-10B3-4, 27-29B3-4,47-^9B3-4, 66-68B3-4, 87-89B3-4, 107-109B3-4, 128-130B3-5, 0-2B3-5, 21-23B3-5, 4 2 ^ 4D3-1, 70-72D3-1,90-92D3-1, 110-112D3-1, 130-132D3-2, 10-12D3-2, 30-32D3-2, 50-52D3-2, 70-72D3-2, 90-92D3-2, 110-112D3-2,130-132D3-3, 10-12D3-3, 30-32D3-3, 50-52D3-3,70-72D3-3, 90-92D3-3, 110-112D3-3, 130-132D3-4, 11-13D3-4, 31-33D3-4, 51-53D3-4, 71-73D3-4, 91-93D3-4, 111-113D3-4, 131-133D3-5, 10-12D3-5, 30-32D3-5, 50-52D3-5, 70-72D3-5, 110-112D3-5, 130-132D3-6, 10-12D3-6, 30-32D3-6, 50-52D3-6, 70-72D3-6, 90-92D3-6, 110-112D3-6, 130-132D3-7, 10-12D3-7, 30-32D3-7, 50-52D3-7,70-72C4-1,48-50C4-1,68-70

Depth(mbsf)

14.9415.1415.3415.5415.7416.0416.2416.4416.6416.8417.0417.1817.3817.5817.7817.9818.1818.3818.5918.7918.9919.1919.3919.5919.7920.1020.2920.4920.6920.8921.0921.2921.0921.2821.4821.6721.8822.0822.2922.5122.7222.9323.7123.9124.1124.3124.6124.8125.0125.2125.4125.6125.8126.1126.3126.5126.7126.9127.1127.3127.6227.8228.0228.2228.4228.6228.8229.1129.3129.5129.7130.1130.3130.6130.8131.0131.2131.4131.6131.8132.1132.3132.5132.7131.4931.69

Rev

mcd

16.8517.0517.2417.4417.6417.9318.1318.3318.5418.7418.9419.0819.2819.4819.6819.8820.0820.2820.4820.6820.8821.0921.2921.4821.7022.0022.2022.3922.5822.7622.9523.1523.1823.3423.5123.7023.9324.1324.3824.5424.7524.9725.3225.4225.4825.5526.0226.3526.4626.6027.1027.3927.5127.7728.0128.2228.4128.6128.8129.0229.3129.5229.7329.9130.1230.3230.5130.8131.0131.2131.4131.8132.0532.4932.6832.8433.0533.2833.4133.5033.7434.1734.3834.5834.6934.89

Age

438.6443.7448.3452.6456.6461.9465.3468.7472.5476.3480.4483.7488.8494.7501.2507.9514.4520.6525.6529.4534.2542.2552.1561.0569.0577.8583.5589.2595.7603.0611.8621.5623.0629.7635.3640.1644.9648.5652.6655.2658.9663.5674.6679.2682.3686.2710.2720.0722.9726.4739.5747.6750.9757.9764.0769.0773.3111.1782.0786.2791.5795.4799.5803.3808.0812.9818.0826.8833.3840.2847.2860.8868.0879.4883.8887.4892.0897.1899.9901.6905.5912.5917.1922.5925.9932.8

LSR

3.84.04.44.85.35.65.95.75.45.14.64.03.63.23.03.03.13.64.54.73.22.32.12.4

3.13.43.43.12.72.32.12.12.32.73.54.4

5.15.86.15.95.23.62.92.11.91.92.43.53.93.93.73.63.73.84.14.34.54.64.85.35.45.34.94.64.33.93.53.33.02.92.93.13.74.04.4

4.54.54.64.95.96.15.54.13.53.02.7

WBD

1.571.551.581.551.571.541.461.451.511.511.521.521.49.52

1.511.50

.511.481.381.431.421.521.491.441.441.431.461.501.471.471.481.461.461.501.481.471.471.461.471.461.431.42

.401.391.431.441.451.441.471.471.441.461.441.421.361.36

.40

.47

.51

.47

.45

.46

.45

.46

.41

.41

.401.401.391.451.41

.46

.491.451.471.481.501.491.471.481.491.461.421.40

.38

DBD

0.700.680.710.670.700.650.550.540.610.620.630.630.590.630.610.600.620.570.440.500.500.640.590.520.520.510.550.600.570.560.570.550.550.600.570.560.570.550.550.550.500.490.470.450.510.530.540.520.560.560.520.550.520.490.410.410.460.560.610.560.530.540.530.540.480.480.470.470.450.530.480.550.550.590.540.560.580.600.580.560.580.590.550.500.460.44

BAR

2.62.73.13.23.73.73.23.13.43.12.92.62.12.01.91.81.92.02.12.51.71.41.21.3

.61.81.9.9.5

1.31.21.21.21.62.02.43.03.23.43.32.72.11.10.9

.00.91.52.02.22.21.92.01.9

.9

.7

.82.12.62.92.92.92.92.62.52.01.91.71.51.31.51.41.71.92.52.42.52.62.72.93.03.93.02.21.71.4

.2

MeanTC

9.910.09.89.99.67.79.0

10.18.88.28.17.79.59.59.08.98.49.08.29.09.48.28.66.77.07.98.37.5

6.96.47.87.88.68.78.78.98.48.78.27.67.40.65.87.88.27.37.77.4

7.36.87.45.64.14.85.77.88.58.06.56.06.27.78.2

7.45.85.27.96.16.98.09.38.89.18.99.68.58.77.88.98.27.07.16.4

CaCO3

76.480.179.079.777.561.773.0S0.570.663.163.961.176.376.272.571.467.373.065.371.868.655.467.852.853.562.665.457.9

54.050.853.559.861.467.767.570.566.068.265.060.156.347.940.658.663.556.357.555.5

55.551.255.640.928.334.042.061.967.563.650.144.848.461.064.9

58.044.839.861.646.754.063.373.469.371.971.176.668.070.561.169.465.054.556.751.2

CaCO3-AR

2.02.2

2.52.62.92.32.42.52.42.01.81.61.61.61.31.31.31.51.41.81.10.80.80.70.91.11.31.1

0.70.60.60.71.01.41.62.12.12.32.11.61.20.50.40.60.60.81.11.2

1.11.01.00.80.50.60.91.62.01.81.51.31.31.51.3

1.00.70.50.90.60.91.21.81.61.81.92.12.02.12.4

2.11.40.90.80.6

Mean

TOC

0.760.350.270.370.310.240.280.430.300.580.390.340.300.370.280.340.300.280.330.390.420.610.440.360.560.410.450.530.490.450.320.320.650.500.570.570.490.510.510.360.410.610.860.950.760.600.530.750.710.400.610.610.770.680.680.730.620.410.430.360.460.400.400.400.370.500.420.460.430.500.490.430.440.450.480.430.400.380.350.240.420.530.350.420.290.22

±lσ

0.340.060.060.060.020.020.070.010.010.140.010.000.010.060.050.030.030.080.06

0.150.480.120.000.030.060.070.020.010.020.010.100.050.080.030.010.010.020.000.000.000.040.010.010.000.020.020.010.01

0.000.030.080.020.040.190.000.000.030.100.010.07

0.030.030.020.000.000.010.030.020.070.060.030.010.010.040.000.010.000.010.020.000.010.000.01

TOC-AR

19.99.48.4

11.911.49.19.0

13.310.218.211.1

8.96.47.65.36.25.85.66.99.97.0

8.65.34.59.07.48.79.97.55.83.83.7

1.18.1

11.613.914.516.417.311.711.112.7

9.68.7

7.35.67.8

14.615.78.9

11.612.114.412.711.513.113.010.612.410.513.611.510.410.17.59.37.06.95.87.76.77.4

8.511.111.310.610.610.510.07.2

16.615.8

7.87.24.22.6

u 3

K

7

0.7630.7560.7550.7190.7090.758

0.7500.8320.7840.6450.800

0.8360.7610.8330.803

0.8270.7820.7850.9670.8360.807

0.9000.8130.8230.8380.8470.8270.9110.7910.7940.8270.8180.8020.8120.8350.8470.8450.8490.8600.8020.809

0.8420.8290.7950.8020.834

0.8150.7630.865

0.8260.8110.8650.7650.7660.7350.7540.8320.826

0.8510.8780.8360.804

0.7350.7820.8070.8130.7920.8780.7530.7800.7550.8380.900

SST ±lσ

21.321.121.120.019.721.1

20.923.321.917.822.4

23.421.223.422.5

23.221.921.927.323.422.6 0.35

25.322.823.123.523.823.225.622.122.223.222.922.422.723.423.823.723.824.122.422.6

23.623.222.2 0.1422.423.4

22.821.324.3 0.07

23.122.724.321.421.420.521.023.323.1 1.70

23.924.723.422.5

20.521.9 0.0722.622.822.124.721.021.821.123.5 0.0025.3

608

ALKENONE SEA-SURFACE TEMPERATURES

Table 1 (continued).

Core, section,interval (cm)

C4-1,89-91C4-1, 110-112C4-1,130-132C4-2, 2-4C4-2, 22-24C4-2,42^44D4-1, 10-12C4-2, 62-64D4-1,30-32C4-2, 83-85D4-1, 50-52D4-1, 70-72D4-1, 90-92D4-1, 110-112D4-1, 130-132D4-2, 10-12D4-2, 30-32D4-2, 50-52D4-2, 70-72D4-2, 90-92D4-2, 110-112D4-2, 130-132D4-3, 10-12D4-3, 30-32D4-3, 50-52D4-3, 70-72D4-3, 90-92D4-3, 110-112D4-3, 130-132D4-4, 10-12D4-4, 30-32D4-4, 50-52D4-4, 70-72D4-4, 90-92D4-4, 110-112D4-4, 130-132D4-5, 10-12D4-5, 30-32D4-5, 50-52D4-5, 70-72D4-5, 90-92D4-5, 100-102D4-5, 120-122D4-5, 140-142D4-6, 10-12D4-6, 30-32D4-6, 50-52D4-6,70-72D4-6, 90-92D4-6, 110-112D4-6, 130-132D4-6, 148-150D4-7, 10-12D4-7, 30-32D4-7,50-52D4-7, 70-72

Depth(mbsf)

31.9032.1132.3132.5332.7332.9332.6133.1332.8133.3433.0133.2133.4133.6133.8134.1134.3134.5134.7134.9135.1135.3135.6135.8136.0136.2136.4136.6136.8137.1137.3137.5137.7137.9138.1138.3138.6138.8139.0139.2139.4139.5139.7139.9140.1140.3140.5140.7140.9141.1141.3141.4941.6141.8142.0142.21

Revmcd

35.1135.3135.5135.7335.9336.1336.3236.3336.5236.5436.7036.9037.0937.2937.4937.7837.9738.1738.3638.5738.7538.9539.2539.4639.6639.8740.1040.3440.5340.7440.8841.0541.2541.5141.7541.9542.2242.4142.6142.8243.0343.1243.3143.5143.7243.9244.1144.2944.4844.6944.9145.1045.2145.3945.5845.80

Age

941.4949.6956.6962.6968.1974.5981.7982.2990.7991.7

1000.01012.01023.01033.01041.01049.01053.01057.01060.01063.01066.01069.01074.01078.01082.01087.01092.01099.01104.01109.01113.01117.01122.01128.01133.01138.01145.01149.01154.01159.01163.01165.01169.01173.01178.01183.01188.01193.01200.01210.01221.01230.01234.01240.01245.01251.0

LSR

2.52.63.23.73.42.92.62.22.21.91.81.71.92.23.14.04.95.66.76.56.36.35.75.14.64.43.93.64.03.93.94.14.24.54.43.94.24.34.14.75.04.74.94.64.13.93.73.12.42.02.12.32.93.43.73.7

WBD

1.491.521.501.461.501.501.491.501.531.481.491.511.501.551.501.521.521.521.501.461.481.461.451.451.461.481.411.391.421.451.441.431.421.421.421.381.381.431.431.421.481.461.481.501.481.491.491.471.451.361.461.481.481.451.431.41

DBD

0.580.630.600.550.600.610.600.600.650.580.590.610.600.670.610.630.630.630.610.550.570.550.540.530.550.570.480.450.500.540.520.500.500.500.500.440.440.510.510.500.570.540.580.600.580.580.590.570.540.400.550.580.570.540.510.48

BAR

1.41.62.02.12.01.71.41.41.31.21.11.01.11.41.82.83.33.63.73.53.73.43.02.72.62.41.91.71.92.10.82.12.12.12.11.81.82.12.22.22.62.52.82.82.62.42.11.71.30.8

• 1.11.41.51.71.92.0

MeanTC

8.28.58.98.59.08.88.2

7.1

8.17.98.46.56.87.37.77.96.06.36.87.86.66.87.47.65.05.86.17.47.67.16.56.66.45.16.77.17.16.26.9

7.98.08.48.3

10.58.78.05.67.17.17.27.38.07.8

CaCO3

66.569.270.368.673.471.765.9

56.7

64.063.168.151.754.158.360.961.247.350.253.861.552.254.659.059.538.544.847.559.361.456.051.049.349.138.952.455.155.147.054.4

61.963.767.165.783.568.762.640.354.956.356.657.061.158.9

CaCO3-AR

1.01.11.41.41.51.31.0

0.8

0.70.70.70.71.01.72.02.21.81.82.02.11.61.51.51.50.70.80.91.30.51.21.11.01.00.70.91.21.21.01.4

1.71.81.81.61.71.10.80.30.60.80.91.01.21.2

MeanTOC

0.210.190.370.270.210.220.270.240.310.230.370.310.250.240.260.340.370.580.300.320.330.420.290.290.300.370.360.410.400.260.240.360.410.670.480.470.580.450.460.560.410.350.420.390.360.420.440.440.520.770.480.300.370.430.640.76

±lσ

0.000.020.030.010.050.010.020.000.010.040.010.060.010.040.010.020.000.050.010.060.080.010.010.020.010.070.020.000.010.030.020.140.020.080.140.030.070.180.000.020.010.000.010.070.100.110.080.010.010.000.070.080.050.060.080.04

TOC-AR

2.93.17.35.74.33.83.83.44.22.64.03.22.73.54.79.6

12.320.911.311.212.114.48.87.87.59.06.97.07.65.52.07.58.7

14.210.28.6

10.59.5

10.112.410.89.0

11.711.19.49.99.37.36.76.45.54.25.77.4

12.215.7

u£0.8840.8060.8770.8560.8590.8470.8160.8590.8210.867

0.7940.8310.8490.8430.8170.791

0.8190.8140.8610.7880.8460.878

0.8410.792

0.8320.8630.850

0.8160.8040.8550.8460.8400.818

0.8530.8670.8720.9570.9270.8960.918

0.8510.8200.8630.8470.9230.9060.910

0.814

SST

24.922.624.624.024.123.822.824.123.024.4

22.223.323.823.622.922.1

22.922.824.222.023.724.7

23.622.1

23.324.223.9

22.922.524.023.723.622.9

23.924.424.527.026.125.225.9

23.923.024.223.826.025.525.6

22.8

±lσ

0.21

0.50

0.070.00

0.07

0.07

0.85

0.21

0.07

0.28

0.35

0.21

Notes: ODP depth in meters below seafloor; rev mcd: revised composite depth (Hagelberg et al., this volume); Age: age assignment (in Ka) for each sample midpoint based on Mixet al. (this volume); LSR: linear sedimentation rate (in cm/k.y.); WBD: GRAPE wet-bulk-density (g/cm3); DBD: calculated dry-bulk density (in g/cm ); BAR: Bulk accumulationrate (in g/cm k.y.~ ); Mean TC: average wt.% total carbon; CaCθ3: wt.% CaCθ3; CaCθ3-AR: calcium carbonate accumulation rate (g/cm k.y.~ ); Mean TOC: average wt.%organic carbon; TOC-AR: organic carbon accumulation rate (mg/cm k.y.~ ); SST: sea-surface temperatures (°C) calculated from U3* indexes.

the weight percentage data for organic carbon and CaCO3, the esti-mated DBD, and the U3

k

? index plotted vs. composite depth. Organiccarbon concentrations increase from the seafloor gradually to 1.5% ataround 1 mcd and slowly decrease to values less than 0.5% at about20 mcd. The gradual decline in TOC concentrations is irregular, andconcentrations vary between 2% and 0.5%. Below 20 mcd, variabilityis lower and values fluctuate from 0.5% to 1%. The CaCO3 concen-trations decrease from the youngest samples near the sediment/waterinterface (75%) to 5 mcd (30%), increase slowly to 80% at 15 mcd,and again decrease to <30% at 27 mcd. Below 15 mcd, the CaCO3

record varies more strongly than above. We note a strong dependenceof DBD on calcium carbonate content; DBD traces both the generalcurve of carbonate and fluctuations on smalW scales. Decreasingdensities in samples from 15 to 3 mcd are attributable to an increasein the diatom content ("Site 846" chapter in Mayer, Pisias, Janecek,et al., 1992), which lowers the sediment density and dilutes CaCO3

concentrations. High SiO2 concentrations correlate negatively with

CaCO3 and do not correlate with A12O3 concentrations in the upper-most cores at Site 846 (Torres, pers. comm., 1993).

Uii indexes plotted vs. depth fluctuate between 0.7 (which corre-sponds to coldest SST estimates) and 1 (which corresponds to warm-est SST estimates) over our record. The interval between 20 and 15mcd, corresponding to highest calcium carbonate concentrations,marks a transition from a cooling to a warming trend toward ouryoungest samples. The warming trend underlies highly variable u i lindexes in the upper record section. It tracks an increase in TOCconcentrations, a decrease in CaCO3 concentrations, and a decreasein DBDs attributed to an increase in opal concentrations.

Accumulation Rates and SSTs

Estimates of accumulation rates rely primarily on linear sedimen-tation rates that are based on the stable oxygen isotope record ofbenthic foraminifers. Sedimentation rates were constrained to vary

609

K.-C. EMEIS, H. DOOSE, A. MIX, D. SCHULZ-BULL

TOC (wt %)

1.0 2.0

CaCO3(wt%)

30 50 70 90

Dry-bulk density (gfcm3) ^ U* 37

0.3 0.4 0.5 0.6 0.7 0.7 0.8 0.9 1.0

i i i

I I I I I I I I I I I I I I I

Figure 3. A. Concentrations of organic carbon. B. Concentrations of calcium carbonate. C. DBD estimates. D. U3

k

? indexes in the top 46

m revised depth in the sediment at Site 846.

smoothly (Mix et al., this volume), and variance of LSR and accumu-lation rates at periodicities shorter than 40 k.y. may have been sup-pressed. Our accumulation rate calculations are further based onestimated sediment DBDs in the composite section, not on actualmeasurements of grain densities. In consequence, they must be re-garded as preliminary.

The percentage data of CaCO3 and TOC at Site 846 translate intoaccumulation rate (AR) curves of considerable variability (Fig. 4).CaCO3-ARs in the older section (from 1300 to 500 Ka and equivalentto 20 mcd), vary between 0.5 and 2.5 g/cm2 k.y."1 without any under-lying trend. After a maximum of CaCO3-accumulation between 400and 500 Ka (2 to 3 g/cm2 k.y."1.), the rates decrease significantly tolowest rates (0.2 mg/cm2 k.y."1) at 100 Ka. This decrease reflects thedecline in concentrations from 15 to 5 mcd. Pronounced fluctuationsrelated to glacial/interglacial cycles (depicted in the δ 1 8 θ record) char-acterize the entire record. These fluctuations are most pronounced inthe lower section of the record to 500 Ka, but are also superimposedon the decreasing trend after this time.

TOC-ARs in the record segment older than 500 Ka range from 3to 20 mg/cm2 k.y."1. After this time, the time series of TOC accumu-lation has a clear trend toward higher rates in younger samples. Inaddition, the rates of TOC burial show increased scatter since 480 Ka,with high variability (between 40 and 5 mgC/cm2 k.y.~1) lasting intothe Holocene.

In a series of cores spanning the entire central and eastern equato-rial Pacific Ocean from the Holocene to 300 Ka, Lyle et al. (1988)noted intense maxima of TOC accumulation at periods centered atabout 18, 150, and 280 Ka. The maximum in isotope stage 2 is notwell represented in our record and blends into a period of high TOC

accumulation during isotope stages 2 through 4, but we can confirmthat the maxima at 150 and 250 Ka are among the strongest and bestdefined in the entire period since 1300 Ka. Both are accompanied byhigh rates in carbonate accumulation.

Calcite accumulation rates in core tops are between 0.9 and 1.2 gCaCO3/cm2 k.y."1 and TOC accumulation rates are about 10 mg/cm2

k.y."1 in the vicinity of Site 846 (Lyle, 1992). Comparing the accumu-lation rate records with the record of δ 1 8 θ in benthic foraminifers(Fig. 4), we noted a good match in the timing of maxima in accumu-lation of biogenic material with increases in δ 1 8 θ during glacials.Both carbonate and organic carbon accumulated at more than twicetheir modern rates and those of interglacials in glacial stages. We alsonoted that an increase in the amplitude of the isotope record between500 and 400 Ka coincides with the maximum of carbonate burial andthe onset of high variability in TOC accumulation rates. Preliminarydata indicate that this time coincides with an increase in burial of opalin the sediments at Site 846 (Torres, pers. comm., 1993).

SSTs at Site 846 varied between 27° and 20°C in the entire record(Fig. 4). Short-term temperature variability was considerable andappears to modulate long-term trends. We note that SSTs decreasedfrom an average of 24°C in the deepest part of our record to approxi-mately 500 Ka (22°C), decreased to values below 20°C at 480 Ka, andincreased from that time to values of approximately 25 °C in ouryoungest samples. Temperature variability in the lower record sectionis less than that in the upper portion, except for a phase of highvariability in the period from 1000 to 800 Ka. During the transitionperiod between the two trends, between 500 and 400 Ka, cold tem-peratures coincide with pronounced maxima in the accumulation ofbiogenic material, most notably of carbonate. In the upper portion of

610

" 3löOUvigerina

6 5 4

B TOC accumulatk>π rate

ALKENONE SEA-SURFACE TEMPERATURES

C DCaCOQ-accumulation rate Sea surface temperature

(mg/cm k.y. )

3θ 10 30 0(g/cm2ky.-1)

200

400

£-600

f

800

1000

1200

19 21 23 25 27

I I I I I

Figure 4. A. The benthic δ 1 8 θ (Mix et al., this volume) record. B. Accumulation rates of organic carbon. C. Carbonate accumulation rates.D. SST calculated from U3

k

? indexes at Site 846 plotted vs. time. In the SST curve, filled circles represent data points that were rejectedand not included in the cross-spectral analyses. Circles are temperature estimates based on measurements of Uii on board JOIDESResolution during Leg 138.

the record, variability in SST is high (over a range of 7°C) anddisplays patterns of glacial low and interglacial high SSTs.

We realize that the temperature calibration of the U k index,which is based on the genus E. huxleyi, is likely to be correct only forthe last 260 k.y., corresponding to a depth of approximately 10 mcdat Site 846. The long trend to decreasing temperatures in the lowerportion of our record may be a reflection of species change thatcontributed to the alkenone signal, which results in a drift of the U3

k

?.However, Marlowe et al. (1990) showed that there is reason to believethat the metabolic function of the alkenones as cell-wall lipids extendsmuch farther into the Phylogenetic lineage of the prymnesiophytes.Based on the correlation between low unsaturation indexes and highCaCO3 concentration in the period between 400 and 500 Ka, corre-sponding to a depth interval between 20 and 15 mcd, we can arguethat the temperature low may indeed indicate colder temperatures insurface water, and that this water at this time was more productive forprymnesiophytes (coccolithophores). We see this as an indication thatthe inflection of the temperature record is not related to changes in themetabolism of alkenone-producing organisms.

DISCUSSION

Long-term Trends

East-west elongated contours of biogenic accumulation rates inmodern sediments (Lyle, 1992) and of SSTs reconstructed from U3

k

?

ratios in surface sediments (Fluegge, unpubl. data) depict zonal gra-

dients that are weak compared to the steep north-south gradients.Back-tracking the relative plate motion of Nazca Plate showed thatSite 846 moved only approximately 92 km due east and parallel to theprevailing direction of modern accumulation rate contours during the1.3 m.y. covered by our record (Mayer, Pisias, Janecek, et al., 1992).The strong change in the character of accumulation rates and SSTcurves between 500 and 400 Ka thus are unlikely to result from sitewandering underneath any permanent hydrographic fronts at the seasurface; they most probably represent a change in long-term condi-tions at the sea surface over the site.

The record segment of temperature inversion to higher tempera-tures, coupled with a change in accumulation rate patterns (increasingTOC and decreasing CaCO3 accumulation rates) between 500 and400 Ka, may be related to the so-called Mid-Brunhes Event. Accord-ing to Jansen et al. (1986), the period between 400 and 300 Ka saw anorthward displacement of oceanic fronts that resulted from an inten-sification of Southern Hemisphere circulation in all ocean basins.However, it appears that the event occurred earlier at Site 846 and thatits effect on sea-surface conditions at Site 846 differed significantlyfrom that of the Central Pacific Ocean.

Our record is similar to that described by Schramm (1985) for twocores (VI9-29 and RC 10-65) from the eastern equatorial Pacific.Schramm (1985) found that radiolarian faunas indicative of coldwater (her "Subpolar Assemblage") had decreased systematicallysince approximately 500 Ka, which she interpreted as a relativedecline in the influx of cool, subpolar waters and a warming trend in

611

K.-C. EMEIS, H. DOOSE, A. MIX, D. SCHULZ-BULL

the eastern equatorial Pacific. This warming trend is well representedin our SST data. At the same time, Schramm (1985) found thatradiolarian fauna indicative of coastal upwelling and high fertilitywaters increased steadily from 300 to 50 Ka in Core V19-29 (3°35'S,83°56'W), which is close to Site 846. Farther west and on the equator,radiolarians in Core RC10-65 (0°41'N, 108°37'W) recorded a de-crease in the importance of upwelling to approximately 400 Ka,followed by a steady increase to 150 Ka and a decrease in upwellingto the present (Schramm, 1985). This pattern of increasing fertility isrepresented at Site 846 by an increase in TOC accumulation, a de-crease in CaCO3 accumulation, and a decrease in dry sediment densi-ties that are attributable to increasing opal contents. Since approxi-mately 120 Ka (5 mcd), increasing sediment densities and increasingCaCO3 accumulation are evident in our record. This can be an indica-tion for decreasing fertility of the surface waters over Site 846 andappears to be consistent with the decrease in faunal upwelling indica-tors, both at the equator and in the western propagation of the PeruCurrent (Schramm, 1985).

The faunal and eolian grain-size data set of Core RC11-210(1°49'N, 140°03'W; Pisias and Rea, 1988) contains a long record (to850 Ka) of the environment in the central equatorial Pacific. Here, amajor change in the record character occurred at 300 Ka and is attrib-uted to the mid-Brunhes Event. High amplitude variations before, anda reduction in amplitude after, 300 Ka in sediments of the central equa-torial Pacific are a mirror image of the SST and accumulation-rate timeseries at Site 846, which display a marked increase in variabilitytoward the top. In addition, the change in record character of both SSTsand accumulation rates occurs approximately 100 k.y. earlier at Site846 than in the central Equatorial Pacific. The reasons for differencesin timing and effect of the event between 400 and 500 Ka between thecentral and eastern equatorial Pacific are as yet unclear. Analyses ofother long records from the equatorial Pacific will help to clarify ifthese are indeed the same events, if they are propagating from the eastto the west, and why their effect is so drastically different nearer tothe ocean margin.

Short-term Variability

Our preliminary analysis of temporal patterns (frequencies andphase relationship with δ 1 8 θ) in SST and biogenic accumulation ratesis limited by the resolution of our records and by the underlying agemodel: for cross-spectral analyses with the δ 1 8 θ record, SST andaccumulation rate records were smoothed with a 24-k.y. Gaussianfilter to remove periods <12 k.y.; these smoothed records were line-arly detrended and sampled at 8-k.y. intervals over the period from 0to 1230 k.y. The records show no coherency in the precession band;however, this could reflect partial attenuation of this frequency by theGaussian smoothing. Table 2 gives the resulting estimates of coher-ency and phase between the temperature and accumulation rate rec-ords in relation to the - δ 1 8 θ record. Conventionally, a phase of 0° inthe individual frequency bands of eccentricity (1/100 k.y.) and obliq-uity (1/41 k.y.) corresponds to maxima of the parameter associatedwith maximal interglacial conditions. In our case, this parameter is- δ 1 8 θ of Mix et al. (this volume). A phase of ±180° means maximumat glacial extremes of the parameter tested vs. - δ 1 8 θ ; negative phaseleads interglacials, positive phase lags interglacials.

The temperature record is coherent with - δ 1 8 θ over the frequencybands of eccentricity (1/100 k.y.) and of obliquity (1/41 k.y.). However,spectral variance is weak at the frequency of obliquity, as it is in thetime series of accumulation rates. In the eccentricity cycle, phaserelationships between the SST and isotope records indicate that warm-est temperatures lead interglacial events and minimal global ice vol-ume by 7 ± 5 k.y. This lead is also seen in the obliquity cycle, wherewarm SSTs lead - δ 1 8 θ by 5 ± 2 k.y.

The time series of organic carbon accumulation is also coherentwith - δ 1 8 θ in the eccentricity cycle and its phase corresponds tohighest accumulation rates 8 k.y. after glacial maxima of ice volume.

Table 2. Summary of cross-spectral analyses.

Cycle100 k.y.

Coherency100 k.y.Phase

41 k.y.Coherency

41 k.y.Phase

-δ 1 8 θvs . SST 0.80 -25° ±18° 0.86 -43° ±15°- δ 1 8 θ vs. CaCO3-AR n.c. 0.91 -41° ±11°- δ 1 8 θ vs. TOC-AR 0.81 -153° ±18° 0.77 HE ±21°

Notes: Time interval is 0-1.23 Ma. n = 155, t = 8 k.y., 50 lags. The 80% confidence intervalfor coherency is 0.634, bandwidth is 0.0033 k.y.' . The figures give phase andcoherency estimates relative to -δ O. n.c. = not coherent.

Organic carbon accumulation is also coherent with - δ 1 8 θ in the obliq-uity cycle; in this band, TOC accumulation occurs 3 k.y. before maxi-mal glacial conditions. The phase relationship of SST and TOC accu-mulation to - δ 1 8 θ in the obliquity cycle suggests that maximal TOCaccumulation occurs coeval with minimal SST in this cycle. Carbon-ate accumulation is not coherent with-δ I 8O in the eccentricity cycleand the spectral density of the obliquity cycle is only weak. However,we note similar phase relationships between CaCO3 accumulationand SST over this band, which implies that carbonate accumulationis maximal at warm temperatures in the obliquity cycle.

At this preliminary stage of our analysis, it appears that on glacial/interglacial time scales, low equatorial SSTs are not linked to in-creased productivities at Site 846, because timing of minima in SSTand maxima in organic carbon burial have been shifted by approxi-mately 15 k.y. in the coherent 100-k.y. eccentricity cycle. In a pre-vious investigation, Lyle et al. (1992) used spectral and cross-spectralanalyses of insolation, δ 1 8 θ and SST, and accumulation rate timeseries to elucidate the relationship between climate-forcing mecha-nisms and timing of maxima in accumulation in relation to upwellingor advection of cold waters in the central equatorial Pacific. Animportant result of their investigation was that the SSTs there werecoherently tied to, but led (5-10 k.y.), global ice volume changes inall primary orbital frequency bands. Our record shows a similar SSTcoherency with, and lead, over global ice volume in the eccentricityband. The lead of SST over global ice volume is consistent with datafrom the tropical Atlantic (Imbrie et al., 1989) and appears to be acharacteristic and pronounced feature of currents emanating from theeastern boundary currents of the Southern Hemisphere (Mclntyre etal., 1989). It reflects an asymmetry in climate change between the twohemispheres and is most likely related to variations in velocity oftrade winds and heat advection from high southern latitudes.

The shift in the eccentricity cycle suggests that high productivityis not coeval with minimal temperatures. In our opinion, this arguesagainst periodically intensified upwelling (which should result in lowSST) as a dominant factor for organic carbon flux to the sediments.Pedersen (1983), Lyle et al. (1988), and others have proposed thatglacial increases in organic carbon burial in the Equatorial Pacificreflect changes in the productivity of surface waters, which are com-monly attributed to intensified upwelling during glacial stages (e.g.,Muller et al., 1983). This intensified upwelling should be reflected insynchronous negative temperature excursions, which are not evidentin our record. Based on the temporal shift in the pronounced eccen-tricity cycle, Lyle et al. (1992) argued that productivity and biogenicparticle flux may be unrelated to upwelling and cold-water advection.Our data agree with this and we propose that the flux of eolianmaterial may be a key factor for determining organic carbon accumu-lation (but not calcium carbonate accumulation) at Site 846. Observa-tions from productivity patterns in the equatorial Pacific (Chavez etal., 1990), sediment trap data (Ittekkot et al., 1992), and numerousdeep-sea cores suggest that organic carbon production, its transferthrough the water column, and its accumulation in the sediment areintimately tied to the flux of eolian aluminosilicate material. Wepropose that increased burial rates of organic carbon during glacialsmay be in part an effect of increased eolian influx into the area atglacial maxima (as proposed for nearby Core VI9-29 in the lateQuaternary by Boyle, 1983). Increased eolian influx may have pro-

612

ALKENONE SEA-SURFACE TEMPERATURES

vided micronutrients for organic carbon production, ballast for a rapidtransfer through the water column, and a seal against remineralizationin the sediment (Davidson et al., in press).

In the shorter 41-k.y. cycle of obliquity, there is indication of adirect link between TOC accumulation and colder, possibly nutrient-rich surface waters. The signal of the obliquity cycle is weak in alltime series, however, and clearly is not as important as the processvarying at the periodicity of 100 k.y.

CONCLUSIONS

Our reconstruction of biogenic accumulation rates and SSTs haveshown that sea-surface conditions at the transition from the PeruCurrent to the SEC of the eastern equatorial Pacific have fluctuatedstrongly over the past 1.3 m.y. A long trend of gradual decrease inSSTs from 24°C at 1.3 Ma ends between 500 and 400 Ka, whenlowest values of 19°C were reached. Since this time, the temperaturedata indicate a warming trend to the Holocene modulated by high-amplitude variation (19°-27°C). The inversion of the trend between400 and 500 Ka coincides with maximal accumulation rates of car-bonate, which since then have decreased. In contrast, organic carbonaccumulation has since then increased in variability and in absolutevalues. On shorter time scales, conditions at the sea surface reflectstrong links to glacial/interglacial cycles over much of the Pleistoceneera. In general, SST is high and productivity is low during inter-glacials. However, differences in their timing with respect to the δ 1 8 θrecord of benthic foraminifers (Mix et al., this volume) suggest thattemperature variations on glacial/interglacial time scales are notimmediately linked to productivity and/or particle flux.

ACKNOWLEDGMENTS

We thank all participants of Leg 138 for a stimulating cruise,N. Pisias, F. Prahl, S. Stewart, and K. Winn for thorough reviews,and the Deutsche Forschungsgemeinschaft (DFG EM-37) for fund-ing our research.

REFERENCES*

Boyle, E.A., 1983. Chemical accumulation variations under the Peru Currentduring the past 130,000 years. J. Geophys. Res., 88:7667-7680.

Brassell, S.C., Eglinton, G., Marlowe, I.T., Pflaumann, U., and Sarnthein, M.,1986. Molecular stratigraphy: a new tool for climatic assessment. Nature,320:129-133.

Bryden, H.L., and Brady, E.C., 1985. Diagnostic model of the three-dimensionalcirculation in the upper Equatorial Pacific Ocean. J. Phys. Oceanogr,15:1255-1273.

Chavez, E, Buck, K.R., and Barber, R.T., 1990. Phytoplankton taxa in relationto primary productivity in the equatorial Pacific. Deep-Sea Res. Part A,37:1733-1752.

Chavez, F.P., and Barber, R.T., 1987. An estimate of new production in theequatorial Pacific. Deep-Sea Res. Part A, 34:1229-1243.

Dandonneau, Y, and Eldin, G., 1987. Southward extent of chlorophyll-enriched waters from the Peruvian and equatorial upwellings betweenTahiti and Panama. Mar. Ecol. Prog. Sen, 38:283-294.

Davidson, E., Ågren, G., Daniel, O., Emeis, K.-C, Largeau, C, Lee, C,Mopper, K., Oades, J., Reeburgh, B., Schimel, D., et al., in press. What arethe physical, chemical, and biological processes that control the formationand degradation of nonliving organic matter? In Zepp, R., and Sonntag, C.(Eds.), The Role of Nonliving Organic Matter in the Earth is Carbon Cycle:Chichester (Wiley).

Eglinton, G., Bradshaw, S.A., Rosell, A., Sarnthein, M., Pflaumann, U., andTiedemann, R., 1992. Molecular record of sea surface temperature changes

on 100-year time scales for glacial terminations I, II, and IV. Nature,356:423-^26.

Halpern, D., Ashby, H., Finch, C, and Robles, J., 1990. TOGA CD-ROMdescription. JPLPubl, 90-43.

Imbrie, J., Mclntyre, A., and Mix, A., 1989. Oceanic response to orbital forcingin the late Quaternary: observational and experimental strategies. In Ber-ger, A., Schneider, S., and Duplessy, J.C. (Eds.), Climate and Geo-Sci-ences: Dordrecht (Kluwer Academic), 121-164.

Ittekkot, V., Haake, B., Bartsch, M., Nair, R.R., and Ramaswami, V., 1992.Organic carbon removal in the sea: the continental connection. In Sum-merhayes, CP., Prell, W.L., and Emeis, K.-C. (Eds.), Upwelling System:Evolution Since the Early Miocene. Geol. Soc. Spec. Publ. London,167-176.

Jansen, J.H.F., Kuijpers, A., andTroelstra, S.R., 1986. Amid-Brunhes climaticevent: long-term changes in global atmospheric and ocean circulation.Science, 232:619-622.

Lyle, M., 1992. Composition maps of surface sediments of the eastern tropicalPacific Ocean. In Mayer, L., Pisias, N., Janecek, T., et al., Proc. ODP, Init.Repts., 138 (Pt. 1): College Station, TX (Ocean Drilling Program), 101-115.

Lyle, M., Murray, D.W., Finney, B.P., Dymond, J., Robbins, J.M., andBrooksforce, K., 1988. The record of late Pleistocene biogenic sedimenta-tion in the eastern tropical Pacific Ocean. Paleoceanography, 3:39-59.

Lyle, M.W., Prahl, EG., and Sparrow, M.A., 1992. Upwelling and productivitychanges inferred from a temperature record in the central equatorialPacific. Nature, 355:812-815.

Marlowe, I.T., Brassell, S.C., Eglinton, G., and Green, J.C, 1990. Long-chainalkenones and alkyl alkenoates and the fossil coccolith record of marinesediments. Chem. Geol., 88:349-375.

Mayer, L., Pisias, N., Janecek, T, et al., 1992. Proc. ODP, Init. Repts., 138(Pts. 1 and 2): College Station, TX (Ocean Drilling Program).

Mclntyre, A., Ruddiman, WE, Karlin, K., and Mix, A.C., 1989. Surface waterresponse of the equatorial Atlantic Ocean to orbital forcing. Paleoceanog-raphy, 4:19-55.

Muller, PJ., Erlenkeuser, H., and von Grafenstein, R., 1983. Glacial-inter-glacial cycles in oceanic productivity inferred from organic carbon con-tents in eastern North Atlantic sediment cores. In Suess, E., and Thiede, J.(Eds.), Coastal Upwelling: Its Sediment Record (Pt. B.): New York (Ple-num), 365-398.

Pedersen, T.E, 1983. Increased productivity in the eastern equatorial Pacificduring the last glacial maximum (19,000 to 14,000 yr B.P.). Geology,11:16-19.

Pisias, N., and Rea, D.K., 1988. Late Pleistocene paleoclimatology of thecentral equatorial Pacific: sea surface response to the southeast tradewinds. Paleoceanography, 3:21-37'.

Prahl, EG., Muehlhausen, L.A., and Zahnle, D.L., 1988. Further evaluation oflong-chain alkenones as indicators of paleoceanographic conditions. Geo-chim. Cosmochim. Acta, 52:2303-2310.

Schramm, C.T., 1985. Implications of radiolarian assemblages for the LateQuaternary paleoceanography of the eastern equatorial Pacific. Quat. Res.(N.Y.), 24:204-218.

Wooster, W.S., 1969. Equatorial front between Peru and Galapagos. Deep-SeaRes. Part A, 16:407^19.

Wyrtki, K., 1981. An estimate of equatorial upwelling in the Pacific. J. Phys.Oceanogr., 11:1205-1214.

Abbreviations for names of organizations and publication titles in ODP reference listsfollow the style given in Chemical Abstracts Service Source Index (published byAmerican Chemical Society).

Date of initial receipt: 1 February 1993Date of acceptance: 31 December 1993Ms 138SR-131

613